categorize_time_operator

Operator to aggregate column features based on a column

class tasrif.processing_pipeline.custom.categorize_time_operator.CategorizeTimeOperator(date_feature_name='date', category_definition='day')

Given a 2D dataframe representing a timeseries where each row represents a time event, this operator will add a new feature(s) that represent a categorization of the date. The categorization specification is provided in the constructor.

Examples

>>> import numpy as np
>>> import pandas as pd
>>>
>>> from tasrif.processing_pipeline.custom import CategorizeTimeOperator
>>>
>>>
>>> dates = pd.date_range('2016-12-31', '2017-01-08', freq='D').to_series()
>>> df = pd.DataFrame()
>>> df["Date"] = dates
>>> df['Steps'] = np.random.randint(1000,25000, size=len(df))
>>> df['Calories'] = np.random.randint(1800,3000, size=len(df))
>>>
>>> df
            Date        Steps   Calories
2016-12-31      2016-12-31      5145    2486
2017-01-01      2017-01-01      5018    2344
2017-01-02      2017-01-02      11010   2426
2017-01-03      2017-01-03      9304    2903
2017-01-04      2017-01-04      13490   2283
2017-01-05      2017-01-05      14511   1976
2017-01-06      2017-01-06      18697   2213
2017-01-07      2017-01-07      19204   2185
2017-01-08      2017-01-08      4470    2333

>>> df5 = df.copy()
>>> operator = CategorizeTimeOperator(date_feature_name="Date",
>>>    category_definition=[
>>>         {"day": "weekday", "values": [1, 1, 1, 1, 0, 0, 1]},
>>>         {"month": "in_may", "values": [0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0]}])
>>> operator.process(df5)[0]
            Date        Steps   Calories        weekday in_may
2016-12-31      2016-12-31      5145    2486    0       0
2017-01-01      2017-01-01      5018    2344    1       0
2017-01-02      2017-01-02      11010   2426    1       0
2017-01-03      2017-01-03      9304    2903    1       0
2017-01-04      2017-01-04      13490   2283    1       0
2017-01-05      2017-01-05      14511   1976    1       0
2017-01-06      2017-01-06      18697   2213    0       0
2017-01-07      2017-01-07      19204   2185    0       0
2017-01-08      2017-01-08      4470    2333    1       0

__init__(date_feature_name='date', category_definition='day')

Creates a new instance of CategorizeTimeOperator

Parameters

• date_feature_name (str) – Name of the feature to identify related timestamp series

• category_definition (str, list, dict) –

  Value is one of “day”, “month” or “hijri_month” to categorize based on day of the week, month of the year or hijri month Array of these values if multiple categorizations are desired.:

  [
      "days", "month"
  ]
  

  Array of dictionary customized column names are desired:

  [
      {"days": "day_of_week"},
      {"month", "calendar_month}
  ]
  

  Array of dictionary with mapping if the default categories are to mapped to customized categories.py For example to categorize based on weekday:

  [
      { "days": "weekday", "values": [1, 1, 1, 1, 0, 0, 1]}
      { "month": "winter", "values": [1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1] }
  ]
  

categorize_duration_operator create_feature_operator